JP6493447B2 - Hot metal temperature prediction method, hot metal temperature prediction device, blast furnace operation method, operation guidance device, hot metal temperature control method, and hot metal temperature control device - Google Patents

Description

  The present invention relates to a hot metal temperature prediction method, a hot metal temperature prediction device, a blast furnace operation method, an operation guidance device, a hot metal temperature control method, and a hot metal temperature control device.

  In the blast furnace process in the steel industry, the hot metal temperature is an important management index. In particular, blast furnace operations in recent years are conducted under conditions of low coke ratio and high pulverized coal ratio in order to pursue rationalization of raw fuel costs, and the furnace conditions are likely to become unstable. For this reason, there is a great need for reducing furnace heat variation. On the other hand, since the blast furnace process is operated in a state of being filled with a solid, the heat capacity of the entire process is large and the time constant of response to the operation is long. Also, there is a dead time of several hours before the raw material charged from the upper part of the blast furnace descends to the lower part of the blast furnace. For this reason, for the furnace heat control, it is essential to optimize the manipulated variables of the manipulated variables based on the future furnace heat prediction.

  From such a background, a method for predicting the hot metal temperature by an approach using a physical model and a method for predicting the hot metal temperature by a statistical approach based on past data have been proposed. Specifically, in Patent Document 1, as the former method, the gas reduction rate parameter included in the physical model is adjusted so as to match the current composition of the furnace top gas, and the physical model after parameter adjustment is used. A method for predicting furnace heat is described. Further, in Patent Document 2, as the latter method, a method of extracting operating conditions similar to the current operating conditions from past operating conditions and predicting the hot metal temperature based on the extracted past operating conditions. Have been described. Furthermore, Patent Document 3 describes a method of constructing a statistical model using past data and predicting the hot metal temperature using the constructed statistical model as the latter method.

JP-A-11-335710 JP 2007-4728 A JP 2008-144265 A

  However, the method described in Patent Document 1 is a method of adjusting the calculated value of the physical model to only the absolute value of the most recent data such as the gas utilization rate by adjusting the parameters, and in the next few hours of the hot metal temperature. No mention is made as to whether the prediction accuracy of the amount of change has improved. In order to control the hot metal temperature, it is necessary to predict how much the hot metal temperature will change in the next few hours. For this reason, according to the method of patent document 1, hot metal temperature cannot be controlled accurately.

  On the other hand, in the method described in Patent Document 2, the prediction accuracy of the hot metal temperature is ensured only when there are similar operation conditions, and there is a concern that the prediction accuracy of the hot metal temperature may be lowered for unknown operation conditions. . In addition, the operating conditions of the blast furnace include many operating variables such as the blowing temperature, coke ratio, and blowing flow rate, and the time constant of the hot metal temperature response is long for all the operating variables. Therefore, when predicting the hot metal temperature, it is necessary to consider the time series data of the manipulated variables. However, since the number of combinations of time series data of manipulated variables becomes enormous, time series data with a high degree of similarity does not necessarily exist.

  Further, in the method described in Patent Document 3, since the reaction result information in the furnace such as the amount of solros carbon several hours ago is required as input information, for example, the operation result immediately after switching the coke ratio is in the furnace. In a situation that has not yet appeared in the reaction result, the prediction accuracy of the hot metal temperature may be lowered.

  The present invention has been made in view of the above problems, and its purpose is to accurately measure the future amount of hot metal temperature without using the latest in-reactor reaction results even under unprecedented operating conditions. It is an object of the present invention to provide a hot metal temperature prediction method and a hot metal temperature prediction device that can be well predicted. Another object of the present invention is to provide a blast furnace operation method, an operation guidance device, a hot metal temperature control method, and a hot metal temperature control device capable of accurately controlling the furnace heat.

  The hot metal temperature prediction method according to the present invention is a hot metal temperature prediction method for predicting the hot metal temperature in the blast furnace using a physical model capable of calculating the state in the blast furnace in an unsteady state, and the reducing material ratio and the amount of solution loss carbon And at least one of the relationship between the reducing material ratio and the gas utilization rate, and the relationship between the reducing material ratio and the furnace body heat loss amount, the sensitivity between the values calculated by the physical model and the actual operation data. The parameters related to the gas reduction rate or the furnace heat loss amount in the physical model are optimized so that the sensitivity of the actual values match, and the hot metal temperature in the blast furnace is predicted using the physical model with the optimized parameters Including the step of:

  The hot metal temperature predicting apparatus according to the present invention is a hot metal temperature predicting apparatus for predicting the hot metal temperature in the blast furnace using a physical model capable of calculating the state in the blast furnace in an unsteady state, and the reducing material ratio and the amount of solution loss carbon. And at least one of the relationship between the reducing material ratio and the gas utilization rate, and the relationship between the reducing material ratio and the furnace body heat loss amount, the sensitivity between the values calculated by the physical model and the actual operation data. The parameters related to the gas reduction rate or the furnace heat loss amount in the physical model are optimized so that the sensitivity of the actual values match, and the hot metal temperature in the blast furnace is predicted using the physical model with the optimized parameters It is characterized by providing the means to do.

  A method for operating a blast furnace according to the present invention includes a step of controlling operating variables of the blast furnace according to a hot metal temperature predicted using the hot metal temperature prediction method according to the present invention.

  The operation guidance apparatus according to the present invention provides the operation of the blast furnace by presenting the relationship between the calculated values of the physical model calculated by the hot metal temperature prediction apparatus according to the present invention and the relationship between the actual values of actual operation data. The present invention is characterized by comprising presentation means for supporting the above.

  The operation guidance device according to the present invention is calculated from a physical model whose parameters are optimized by the hot metal temperature prediction device according to the present invention, and includes a temperature distribution including a heat storage zone position and a chemical storage zone position formed in the furnace. And presenting means for supporting the operation of the blast furnace by presenting the reduction rate distribution.

  The hot metal temperature control method according to the present invention is a hot metal temperature control method for controlling the hot metal temperature based on the hot metal temperature predicted by the hot metal temperature prediction method according to the present invention, wherein the physical model in which the parameters are optimized is provided. To predict the future hot metal temperature when the current manipulated variable of the operating variable of the blast furnace is maintained, and to minimize the difference between the hot metal temperature predicted in the prediction step and the target hot metal temperature Determine the appropriate operating amount of the operating variable of the blast furnace including at least one of the blast moisture, the amount of pulverized coal injection, the coke ratio at the top of the furnace, and the air temperature, and the operating variable of the blast furnace according to the determined appropriate operating amount And a control step for controlling.

  The hot metal temperature control device according to the present invention is a hot metal temperature control device that controls the hot metal temperature based on the hot metal temperature predicted by the hot metal temperature prediction device according to the present invention, wherein the physical model in which the parameters are optimized is provided. To predict the future hot metal temperature when the current manipulated variable of the operating variable of the blast furnace is maintained, and to minimize the difference between the hot metal temperature predicted by the predicting means and the target hot metal temperature Determine the appropriate operating amount of the operating variable of the blast furnace including at least one of the blast moisture, the amount of pulverized coal injection, the coke ratio at the top of the furnace, and the air temperature, and the operating variable of the blast furnace according to the determined appropriate operating amount And a control means for controlling.

  The operation guidance device according to the present invention provides a blast furnace by presenting the amount of change in the hot metal temperature predicted by the prediction means provided in the hot metal temperature control device according to the present invention and the amount of change in the hot metal temperature in the actual operation data. It is provided with a presentation means for supporting the operation.

  The operation guidance apparatus according to the present invention is a hot metal predicted when the operating variable of the blast furnace is controlled according to the transition of the future hot metal temperature predicted by the prediction means provided in the hot metal temperature control apparatus according to the present invention and the proper operation amount. It is characterized by providing a presentation means for supporting the operation of the blast furnace by presenting the temperature transition.

  According to the hot metal temperature prediction method and the hot metal temperature prediction apparatus according to the present invention, it is possible to accurately predict the future amount of change of the hot metal temperature without using the latest in-reactor reaction results even under unprecedented operating conditions. . Moreover, according to the operation method of the blast furnace, the operation guidance device, the hot metal temperature control method, and the hot metal temperature control device according to the present invention, the furnace heat can be accurately controlled.

FIG. 1 is a diagram showing input variables and output variables of a physical model used in the present invention. FIG. 2 is a diagram illustrating a Rist model. FIG. 3 is a diagram illustrating an example of the relationship between RAR and ηCO, the relationship between RAR and the amount of solros carbon, and the relationship between RAR and the amount of furnace heat loss. FIG. 4 is a diagram illustrating a result of calculating the relationship illustrated in FIG. 3 using a physical model before parameter optimization. FIG. 5 is a diagram illustrating the Rist model of the actual operation and physical model. FIG. 6 is a graph showing the relationship between the solid temperature before and after parameter optimization and the gas composition fraction. FIG. 7 is a diagram illustrating a result of calculating the relationship illustrated in FIG. 3 using a physical model after parameter optimization. FIG. 8 is a diagram showing the distribution in the furnace height direction of the temperature in the furnace, the ore oxidation degree, and the gas utilization rate calculated by the physical model with optimized parameters. FIG. 9 is a diagram showing the transition of the manipulated variable of the manipulated variable. FIG. 10 is a diagram illustrating the future prediction of the output variable of the physical model. FIG. 11 is a diagram showing the correlation between the predicted value of the hot metal temperature change amount before and after parameter optimization and the actual measurement value. FIG. 12 is a diagram showing a change in the hot metal temperature accompanying the operation of the blast moisture, and a diagram showing the degree of influence on the hot metal temperature per unit operation amount of the blast moisture. FIG. 13 is a diagram showing a predicted transition of the proper operation amount of the blast moisture and the hot metal temperature during the blast moisture operation. FIG. 14 is a diagram showing a comparison result between the guidance operation amount and the operation amount by the actual operator, and a histogram of the hot metal temperature in each case of compliance with guidance and non-observance.

  Hereinafter, a hot metal temperature prediction method, a hot metal temperature prediction device, a blast furnace operation method, an operation guidance device, a hot metal temperature control method, and a hot metal temperature control device according to the present invention will be described with reference to the drawings.

[Configuration of physical model]
First, a physical model used in the present invention will be described.

  The physical model used in the present invention is the reduction of ore as in Reference 1 (Haneda Michiharu et al .: “Investigation of Fire Operation by Blast Furnace Unsteady Model”, Iron and Steel, vol.68, p.2369). It is a physical model that can calculate the in-furnace state in an unsteady state, which is composed of partial differential equations that take into account physical phenomena such as heat exchange between ore and coke and melting of ore.

  As shown in Fig. 1, the main ones that change with time in the boundary conditions given to this physical model (input variables, blast furnace operating variables (also called operating factors)) are the coke ratio at the top of the furnace (hot metal production). Coke weight used for 1 ton), air flow rate (flow rate of air blown into the blast furnace), enriched oxygen flow rate (flow rate of enriched oxygen blown into the blast furnace), air blowing temperature (air blown into the blast furnace) Air temperature), pulverized coal injection amount (weight of pulverized coal used per 1 ton of hot metal production, PCI), and blowing moisture (humidity of air blown into the blast furnace).

Also, the main output variables of this physical model are the gas utilization rate in the furnace (CO ₂ / (CO + CO ₂ ), ηCO), the raw material and gas temperature, the amount of solution loss carbon (the amount of solution loss carbon), the ironmaking speed ( Hot metal production rate), hot metal temperature, furnace heat loss amount (heat amount taken by cooling water when the furnace body is cooled by cooling water), and reducing material ratio (pulverized coal injection amount per ton of hot metal and coke ratio) Sum, RAR).

  Although the time step for calculating the in-furnace state in the unsteady state is arbitrary, in the present invention, the time step is set to 30 minutes. However, the time step is variable according to the purpose, and is not limited to the value of this embodiment. By using this physical model, it is possible to calculate the in-furnace state and the hot metal temperature which change from moment to moment. In the following, the calculated value and the actual measurement value of the hot metal temperature at time step k are indicated as Cal_HMT (k) and Act_HMT (k), respectively.

[About the Rist model]
Next, an outline of a heat balance material balance model (Rist model), which is an important concept in optimizing the parameters of the physical model, will be described.

As shown in FIG. 2, the Rist model is RAR in the blast furnace, the amount of solros carbon (sol loss), ηCO, oxygen intensity, ironmaking speed, furnace heat loss amount (heat loss), hot metal / slag sensible heat, and This model shows the relationship between the sensible heat and is described in detail in Reference Document 2 (Yoichi Ono: “Rist Operation Diagram (I), Iron and Steel, 79 (1993), N618). According to this, it can be seen that if the RAR is increased by increasing the amount of pulverized coal injection under a constant blowing temperature, the hot metal temperature and the furnace heat loss amount will increase. Even if it increases, it can be seen that the RAR rises as it is pulled in. Thus, since the substance balance and the heat balance affect each other, the causal relationship between the cause and effect is not uniquely determined. There. Note that the shaft efficiency in FIG. 2, which shows the degree of dissociation of the theoretical limit of indirect reduction (CO / CO ₂ ratio of the constraints on the reduction of the FeO) in the blast furnace, ItaCO and Sol Los carbon This is an index that is uniquely determined from the quantity.

  Here, the concept of around the P point and around the W point is mentioned as an important concept in the Rist model. Since the operation line L1 of the Rist model has two degrees of freedom of inclination and intercept, one degree of freedom remains even if the RAR that is the inclination of the operation line L1 is determined. That is, where the rotation center is set when the RAR changes is changed depending on the state in the furnace at that time. As extreme examples, there are cases around the P point and around the W point.

  Around the point P is when the operating line L1 rotates around the point P in the figure, and when the RAR increases, the intersection B of the operating line L1 and the line segment AW moves to the point A side, so that the shaft While the efficiency is greatly reduced, the sensible heat exhausted from the lower part of the furnace does not change. That is, the amount of increase in latent heat per unit amount of iron due to the increase in RAR is used for endotherm due to an increase in sensible heat (increase in CO gas fraction) of the W point exhaust gas and an increase in the amount of Sollos carbon.

  On the other hand, “around the W point” means that the operating line L1 rotates with the shaft efficiency being constant with the W point as the center of rotation, and if the RAR is increased, the sensible heat at the lower part of the furnace is significantly increased. Since the gas composition at the W point (W point exhaust gas calorie) is fixed, the latent heat per unit amount of iron due to the increase in RAR is used as sensible heat at the bottom of the furnace. With the above concept in mind, the parameters of the physical model are optimized.

  3A to 3C show the relationship between RAR and ηCO in the actual operation data of a certain blast furnace, the relationship between RAR and the amount of solros carbon, and the relationship between RAR and the amount of heat loss of the furnace body. Moreover, the result of having calculated the relationship shown in FIG. 3 (a)-(c) by the physical model based on the theoretical formula of the reference document 1 using the operation data in the same period is shown in FIG. 4 (a)-(c). Shown in

  As shown in FIG. 3 (b), in the actual operation, since the slope has a negative correlation between the RAR and the amount of solros carbon, the operation line L1 is around the W point. On the other hand, as shown in FIG. 4B, since the slope has a positive correlation between the RAR and the amount of solros carbon on the physical model, the operating line L1 is P on the physical model. It ’s better than around the dot. Such a difference also appears between RAR and ηCO. In the slope of the correlation between RAR and ηCO, the reason why the slope in actual operation is gentler than the slope in the physical model is that the rotation center of the operation line is located higher.

  As a result, the Rist model of actual operation is represented as shown in FIG. 5A, and the Rist model of the physical model is represented as shown in FIG. 5B. That is, as shown in FIG. 5A, in the actual operation, the operation line L1 becomes the operation line L1 'by rotating around the point B as being around the W point. As a result, the RAR increases and the amount of solros carbon decreases. On the other hand, on the physical model, as shown in FIG. 5B, the operation line L1 rotates around the point P (point P ′) by rotating around the point P, thereby operating the operation line L1. L1 ′. As a result, RAR increases and the amount of solros carbon increases.

[Parameter optimization method]
Next, a method for optimizing the parameters of the physical model will be described.

In general, there is a region where the temperature rise of about 900 to 1000 ° C. called a heat preservation zone is stagnant inside the blast furnace, and there is a chemical preservation zone where ore reduction is stagnant inside the heat preservation zone. The gas composition fraction (CO / (CO + CO ₂ )) is considered to reach the Fe—FeO reduction equilibrium line in this region. However, in the physical model before optimizing the parameters, the gas composition does not reach the top of the furnace without reaching the Fe—FeO reduction equilibrium line (CO / CO ₂ equilibrium line) as shown by the broken line in FIG. ing. This means that the gas composition at the W point greatly fluctuates due to the fluctuation of the RAR, and it can be said that the physical model is around the P point.

Originally, the gas composition fraction should reach the Fe-FeO reduction equilibrium line in the temperature range described above, as shown by the solid line in FIG. The gas composition in this temperature range is determined by the balance between the CO consumption reaction by the gas reduction reaction (FeO + CO = Fe + CO ₂ ) and the CO production reaction by the carbon gasification reaction (C + CO ₂ = 2CO). Therefore, in order to make the gas composition fraction reach the Fe—FeO reduction equilibrium line in the temperature range described above, the gas reduction rate is increased (for example, twice the rate) and the carbon gasification reaction rate is decreased (for example, (1/5 speed) is effective. In this example, the parameters on the physical model are set so that the sensitivity (slope of correlation) between the calculated values of the RAR and the solros carbon amount matches the sensitivity between the actual values of the RAR and the solros carbon amount in the actual operation data. Was fixed.

  In addition, the fact that it is around the W point means that the amount of change in the sensible heat of the lower part of the furnace (the amount of heat loss of the furnace body + hot metal temperature) when the unit amount of RAR changes is larger than that around the P point. is there. The hot metal temperature and the amount of heat loss of the furnace body are the same as the amount of heat, and if the RAR is increased, the amount of heat loss of the furnace body naturally increases. That is, the hot metal temperature is determined as a result of competing for the heat generated with the increase in RAR between the furnace heat loss amount and the hot metal temperature. Therefore, in the present invention, the hot metal temperature is predicted after the relationship between the RAR and the furnace heat loss amount is also reflected in the physical model.

  Specifically, in the present invention, the furnace body heat loss amount is given on the physical model so as to reproduce the relationship between the RAR and the furnace body heat loss amount. In this embodiment, the value of the furnace body heat loss amount is set to five times the value obtained from the theoretical formula. In addition, this value was calculated | required so that the sensitivity of the calculated values of RAR and the furnace body heat loss amount might correspond with the sensitivity of the actual values of RAR and furnace body heat loss in actual operation data. The result of optimizing the parameters as described above is shown in FIGS. It can be seen that the tendency toward the actual operation shown in FIGS. 3A to 3C is approached compared to before the parameter optimization shown in FIGS. 4A to 4C. Thus, for example, the operator can grasp the accuracy of the physical model by presenting the relationship between the RAR and the amount of solros carbon in real time for both the actual value and the calculated value. Therefore, it can be an index for judging the reliability of operation guidance for reducing hot metal temperature variation shown below.

  In the present embodiment, the sensitivity between the calculated values based on the physical model and the sensitivity between the actual values of the actual operation data are matched with respect to the relationship between the RAR and the amount of solros carbon and the relationship between the RAR and the amount of heat loss of the furnace body. As described above, the parameters related to the gas reduction rate or the furnace heat loss amount in the physical model were optimized, but the present invention is not limited to this example, and the relationship between the RAR and the amount of solution loss carbon, RAR and As for at least one of the relationship between the gas utilization rate and the relationship between the RAR and the furnace heat loss amount, the sensitivity between the values calculated by the physical model and the sensitivity between the actual values of the actual operation data are matched. In addition, the parameters related to the gas reduction rate or the furnace heat loss amount in the physical model may be optimized.

FIG. 8A to FIG. 8C show distributions in the furnace height direction of the furnace temperature, the ore oxidation degree, and the gas utilization rate calculated by the physical model whose parameters are optimized according to the present invention. The origin of the vertical position of the vertical axis is the tuyere height. The gas utilization rate indicates the rate at which CO gas generated in the furnace is used for ore reduction and converted to CO ₂ gas. As shown in FIG. 8 (a), ore is charged at room temperature at the top of the furnace, and the temperature rises due to heat exchange with the gas as it descends the furnace. The temperature difference between the gas and iron is wide in the region of 2 m or less in FIG. 8 (a). This is because the ore melts and becomes hot metal, and the interface area between the gas and iron phase where heat exchange occurs is reduced. Further, as shown in FIG. 8B, the unreduced ore (FeO1.5) charged at the top of the furnace is reduced by CO gas and H ₂ gas in the process of descending the furnace, so that the oxidation of the ore The degree is lower at the bottom of the furnace. In addition, as shown in FIG. 8C, the gas utilization rate is zero at the tuyere height, and rises by ore reduction in the process in which the gas rises in the furnace.

As already stated, the existence of a heat preservation zone and a chemical preservation zone has been confirmed inside the blast furnace. In this region, the gas utilization rate reaches the Fe—FeO reduction equilibrium. In this physical model, such a region can be reproduced from a height position of 8 to 12 m. This is obtained as a result of adjusting the model so that the CO / CO ₂ ratio reaches the Fe—FeO equilibrium line, as indicated by the solid line in FIG. It is known that the position of the heat preservation zone and the chemical preservation zone moves up and down according to the heat level in the furnace. By presenting such an in-furnace state to the operator, it becomes a reference for daily operation actions. It is expected. For example, when unreduced ore flows into the lower part of the furnace, it leads to a decrease in furnace heat, so that it is possible to take an action to increase the furnace heat before the hot metal temperature decreases.

  Next, a method for determining an appropriate operation amount for reducing the furnace heat variation will be described. Since the heat capacity of the blast furnace process is large, the time constant of the response to changes in the manipulated variables of the blast furnace operating variables is as long as about 12 hours. Therefore, control based on prediction of future in-furnace conditions is effective for reducing furnace heat variation. Therefore, in the present invention, a physical model predictive control system based on the future prediction based on the physical model is constructed. Specifically, physical model predictive control is performed in three steps: prediction calculation of future hot metal temperature assuming that the current operation amount of the operation variables of the blast furnace is maintained, calculation of step response of operation amount, and operation amount optimization. A system was constructed.

First, the prediction calculation of the hot metal temperature in the future is described. Assuming that the current operating quantity of the operating variables of the blast furnace is kept constant in the future, predictive calculation of the future hot metal temperature is performed by repeatedly calculating the physical model. FIGS. 9A to 9D show the future prediction of the manipulated variable of the manipulated variable, and FIGS. 10A to 10D show the future prediction of the output variable of the physical model including the hot metal temperature. The origin of the time is the guidance time point, that is, the timing for making a prediction. 9A to 9D and FIGS. 10A to 10D, the solid line indicates the calculated value, and the broken line indicates the actual measurement value. Actual measured values in the future section that cannot be obtained at the time of guidance are also indicated by broken lines in the figure. As shown in FIGS. 10A to 10D, although the predicted calculation does not reflect the manipulated variable in the future section, the calculated value of the hot metal temperature matches the measured value correspondingly. Thereby, since the heat capacity of the blast furnace process is large, it can be said that the hot metal temperature transition up to 10 hours in the future is greatly influenced by the accumulation of the past manipulated variables. Hereinafter, the calculated value of the hot metal temperature at time t obtained here is defined as free response HMT _free (t).

  Next, the prediction accuracy of the amount of change in hot metal temperature before and after parameter optimization will be described. 11A and 11B show the results of correlation between the predicted value of the hot metal temperature change after 8 hours and the actual value. As shown in FIGS. 11A and 11B, the correlation between the measured value and the calculated value of the hot metal temperature change amount before and after parameter optimization is improved, and the prediction accuracy of the hot metal temperature change amount is improved. It was confirmed. Therefore, according to the present invention, it was confirmed that the future change amount of the hot metal temperature can be accurately predicted without using the latest reaction result in the furnace even under unprecedented operating conditions.

Next, calculation of step response will be described. As shown by broken lines in FIGS. 12 (a) and 12 (b), it is assumed that the operating variable such as blast moisture has been manipulated by a unit amount, and in the same way as when calculating the free response HMT _free (t), the future hot metal temperature transition Calculate Here, the case where the ventilation moisture was changed by 10 g / Nm ³ was shown. 12A and 12B, the solid line indicates the transition of the operation variable when the operation variable is not operated, and the broken line indicates the transition of the operation variable when the operation variable is operated. The operating variables other than the blast moisture are the same as those shown in FIGS. Further, as shown in FIG. 12C, the degree of influence on the hot metal temperature per unit operation amount of the blast moisture is separated by taking the difference from the free response. For other operation variables, the influence degree of the operation variable can be separated by the same procedure.

  In general, in the blast furnace process, the hot metal temperature is controlled to be constant by operating at least one of the blowing temperature, blowing moisture, pulverized coal injection amount, and coke ratio at the top of the furnace. In the following, blast moisture is selected as an operation variable, but the same logic can be constructed for other operation variables.

  Next, a method for determining the optimum operation amount will be described. In general model predictive control, there are two adjustment parameters: a prediction section (how far the section is used as an evaluation function) and a control section (how far the operation amount is optimized). In this embodiment, the prediction interval is 10 hours and the control interval is 1 step. However, these are adjustable values, and are not limited to the values of the present embodiment.

The blast moisture manipulated variable ΔBM for minimizing the evaluation function J shown in the following formulas (1) and (2) consisting of the integral value of the deviation from the hot metal temperature target value HMT _ref and the manipulated variable up to 10 hours ahead is Ask. Here, in the formulas (1) and (2), HMT _pre is a predicted value of the hot metal temperature when the blast moisture is changed, and is obtained by superimposing the effect of the blast moisture on the free response HMT _free (t). is there. Further, a and b are weighting factors. Stp _BM (t) is a step response of the blast moisture. The above problem is reduced to a secondary planning problem. Equations (1) and (2) are for determining the proper operation amount of the blast moisture. Similarly, the proper operation amount of the pulverized coal injection amount, the coke ratio at the top of the furnace, and the blast temperature is also shown. Can be sought.

  FIGS. 13A and 13B show the proper operation amount of the blast moisture determined by the present invention and the predicted transition of the hot metal temperature during the blast moisture operation. 13 (a) and 13 (b), LA is the calculated value of the blast moisture, LB is the actual measurement value of the blast moisture, LC is the guidance value of the blast moisture, LD is the calculated value of the hot metal temperature, and LD 'is the blast. The calculated value of the hot metal temperature when no moisture operation is performed, LE is the actual measurement value of the hot metal temperature, and LF is the actual measurement value of the hot metal temperature when operating the blowing moisture. In this example, the target value of the hot metal temperature was set to 1500 ° C. Thus, it is considered that the excess of the furnace heat can be alleviated by increasing the blown moisture in advance when the excess of the furnace heat can be predicted. Also, by presenting the hot metal prediction transition during no operation and the hot metal temperature prediction transition during guidance operation, it is possible to construct an operation guidance capable of intuitively grasping the influence of the guidance operation.

FIG. 14A shows a result of comparison between the guidance operation amount and the operation amount by the actual operator. The operation amount of the operator was the operation amount from the guidance time point to 2 hours later. The case where the difference in the operation amount between the operation by the operator and the guidance system was 3 g / Nm ^{3 or} less was defined as guidance compliance, and the other cases were defined as non-guidance. FIGS. 14 (b) and 14 (c) show the hot metal temperature histograms when 10 hours have passed since the guidance for each of the cases of compliance with guidance and non-observance. In the guidance compliance case, it was confirmed that the variation in hot metal temperature was reduced by 3.7 ° C. Thus, the effectiveness of the furnace heat control method according to the present invention was verified.

Claims (9)

A hot metal temperature prediction method for predicting the hot metal temperature in a blast furnace using a physical model capable of calculating the state in the blast furnace in an unsteady state,
At least one of the relationship between the reducing material ratio and the solution loss carbon amount, the relationship between the reducing material ratio and the gas utilization rate, and the relationship between the reducing material ratio and the furnace heat loss amount was calculated by the physical model. The parameters related to the gas reduction rate or the furnace heat loss amount in the physical model are optimized so that the sensitivity between the values matches the sensitivity between the actual values of the actual operation data. A method for predicting hot metal temperature, comprising the step of using the method to predict hot metal temperature in a blast furnace. A hot metal temperature prediction device for predicting the hot metal temperature in a blast furnace using a physical model capable of calculating the state in the blast furnace in an unsteady state,
At least one of the relationship between the reducing material ratio and the solution loss carbon amount, the relationship between the reducing material ratio and the gas utilization rate, and the relationship between the reducing material ratio and the furnace heat loss amount was calculated by the physical model. The parameters related to the gas reduction rate or the furnace heat loss amount in the physical model are optimized so that the sensitivity between the values matches the sensitivity between the actual values of the actual operation data. An apparatus for predicting the hot metal temperature, comprising means for predicting the hot metal temperature in the blast furnace.   A method for operating a blast furnace, comprising the step of controlling operating variables of the blast furnace according to a hot metal temperature predicted using the hot metal temperature prediction method according to claim 1.   A presentation means for supporting the operation of the blast furnace is provided by presenting the relationship between the values of the physical model calculated by the hot metal temperature prediction apparatus according to claim 2 and the relationship between the actual values of actual operation data. Operation guidance device characterized by   A temperature distribution and a reduction rate distribution including a heat storage zone position and a chemical storage zone position formed in the furnace, which are calculated from a physical model whose parameters are optimized by the hot metal temperature prediction apparatus according to claim 2, are presented. Thus, an operation guidance device comprising a presentation means for supporting operation of the blast furnace. A hot metal temperature control method for controlling the hot metal temperature based on the hot metal temperature predicted by the hot metal temperature prediction method according to claim 1,
Predicting step of predicting the hot metal temperature in the future when the current manipulated variable of the operating variable of the blast furnace is maintained using the physical model in which the parameter is optimized;
A blast furnace including at least one of blast moisture, pulverized coal injection amount, coke ratio at the top of the furnace, and blast temperature so as to minimize the difference between the hot metal temperature predicted in the prediction step and the target hot metal temperature. A control step of determining an appropriate operation amount of the operation variable of the blast furnace, and controlling an operation variable of the blast furnace according to the determined appropriate operation amount
The hot metal temperature control method characterized by including. A hot metal temperature control device for controlling the hot metal temperature based on the hot metal temperature predicted by the hot metal temperature prediction device according to claim 2,
Predicting means for predicting the hot metal temperature in the future when the current operating amount of the operating variable of the blast furnace is maintained using the physical model in which the parameters are optimized,
A blast furnace including at least one of blast moisture, pulverized coal injection amount, coke ratio at the furnace top, and blast temperature so as to minimize the difference between the hot metal temperature predicted by the prediction means and the target hot metal temperature. A control means for determining an appropriate operation amount of the operation variable of the blast furnace, and controlling the operation variable of the blast furnace according to the determined appropriate operation amount
A hot metal temperature control device comprising:   The presenting means for supporting the operation of the blast furnace by presenting the amount of change in the hot metal temperature predicted by the prediction means provided in the hot metal temperature control device according to claim 7 and the amount of change in the hot metal temperature in the actual operation data. An operation guidance device comprising:   The hot metal temperature transition predicted by the predicting means included in the hot metal temperature control device according to claim 7 and the hot metal temperature transition predicted when the operating variable of the blast furnace is controlled according to the proper operation amount are presented. Thus, an operation guidance device comprising a presentation means for supporting operation of the blast furnace.

Images